[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <2af56bf7-aff8-b4f5-b4cf-06743e906695@huawei.com>
Date: Thu, 14 Nov 2019 14:45:02 +0800
From: Yunfeng Ye <yeyunfeng@...wei.com>
To: "Rafael J. Wysocki" <rjw@...ysocki.net>
CC: <lenb@...nel.org>, <linux-acpi@...r.kernel.org>,
<linux-kernel@...r.kernel.org>, <linux-doc@...r.kernel.org>,
"hushiyuan@...wei.com" <hushiyuan@...wei.com>,
"linfeilong@...wei.com" <linfeilong@...wei.com>
Subject: Re: [PATCH] ACPI: sysfs: Change ACPI_MASKABLE_GPE_MAX to 0x100
On 2019/11/14 7:21, Rafael J. Wysocki wrote:
> On Tuesday, November 5, 2019 3:35:25 PM CET Yunfeng Ye wrote:
>> There are two problems after commit 0f27cff8597d ("ACPI: sysfs: Make
>> ACPI GPE mask kernel parameter cover all GPEs"):
>>
>> 1. ACPI_MASKABLE_GPE_MAX is changed from 0x80 to 0xff, so the check
>> condition "gpe > ACPI_MASKABLE_GPE_MAX" is not valid because the type of
>> gpe is u8.
>>
>> 2. The size of bitmap is ACPI_MASKABLE_GPE_MAX, so it is not support the
>> num 255 for gpe.
>>
>> Update the macro ACPI_MASKABLE_GPE_MAX to 0x100, and change the type of
>> gpe to u32, also modify the checking condition for gpe.
>>
>> Bye the way, update the docs for kernel parameter acpi_mask_gpe.
>>
>> Fixes: 0f27cff8597d ("ACPI: sysfs: Make ACPI GPE mask kernel parameter cover all GPEs")
>> Signed-off-by: Yunfeng Ye <yeyunfeng@...wei.com>
>
> AFAICS this really is about supporting the masking of GPE 0xFF.
>>> ---
>> Documentation/admin-guide/kernel-parameters.txt | 1 +
>> drivers/acpi/sysfs.c | 8 ++++----
>> 2 files changed, 5 insertions(+), 4 deletions(-)
>>
>> diff --git a/Documentation/admin-guide/kernel-parameters.txt b/Documentation/admin-guide/kernel-parameters.txt
>> index a84a83f8881e..dd878e2491e1 100644
>> --- a/Documentation/admin-guide/kernel-parameters.txt
>> +++ b/Documentation/admin-guide/kernel-parameters.txt
>> @@ -114,6 +114,7 @@
>> This facility can be used to prevent such uncontrolled
>> GPE floodings.
>> Format: <int>
>> + Support masking of GPEs numbered from 0x00 to 0xff
>
> Why not just say "Format: <byte>" ?
>
ok, thanks.
>>
>> acpi_no_auto_serialize [HW,ACPI]
>> Disable auto-serialization of AML methods
>> diff --git a/drivers/acpi/sysfs.c b/drivers/acpi/sysfs.c
>> index 75948a3f1a20..3c3302583d78 100644
>> --- a/drivers/acpi/sysfs.c
>> +++ b/drivers/acpi/sysfs.c
>> @@ -819,14 +819,14 @@ static ssize_t counter_set(struct kobject *kobj,
>> * interface:
>> * echo unmask > /sys/firmware/acpi/interrupts/gpe00
>> */
>> -#define ACPI_MASKABLE_GPE_MAX 0xFF
>> +#define ACPI_MASKABLE_GPE_MAX 0x100
>
> So this is the only change that's necessary AFAICS.
>
>> static DECLARE_BITMAP(acpi_masked_gpes_map, ACPI_MASKABLE_GPE_MAX) __initdata;
>>
>> static int __init acpi_gpe_set_masked_gpes(char *val)
>> {
>> - u8 gpe;
>> + u32 gpe;
>>
>> - if (kstrtou8(val, 0, &gpe) || gpe > ACPI_MASKABLE_GPE_MAX)
>
> And here you can drop the ACPI_MASKABLE_GPE_MAX check and the rest can remain
> as is.
>
I will modify as your suggestion, thanks.
>> + if (kstrtouint(val, 0, &gpe) || gpe >= ACPI_MASKABLE_GPE_MAX)
>> return -EINVAL;
>> set_bit(gpe, acpi_masked_gpes_map);
>>
>> @@ -838,7 +838,7 @@ void __init acpi_gpe_apply_masked_gpes(void)
>> {
>> acpi_handle handle;
>> acpi_status status;
>> - u8 gpe;
>> + u32 gpe;
>>
>> for_each_set_bit(gpe, acpi_masked_gpes_map, ACPI_MASKABLE_GPE_MAX) {
>> status = acpi_get_gpe_device(gpe, &handle);
>>
>
>
>
>
>
> .
>
Powered by blists - more mailing lists